The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze and modify Go source code, which serves as an ingestion point for untrusted data. An attacker could potentially embed malicious instructions within the source code of a project to influence the agent's behavior during code reviews or audits.\n
Ingestion points: The skill ingests code via the Read, Glob, and Grep tools as defined in its Audit and Review modes.\n
Boundary markers: The skill relies on its strong "Persona" and technical guidelines to maintain behavior, but does not implement explicit delimiters or "ignore" instructions for the code it processes.\n
Capability inventory: The skill utilizes powerful capabilities including Bash (restricted to go, golangci-lint, and git), Write, Edit, and the Agent tool for sub-task parallelization.\n
Sanitization: There is no evidence of specific sanitization or filtering of the source code content before it is processed by the agent or its sub-agents.

backend-go-error-handling